Bo CAI ,Xiang JI ,Yingyong WANG ,Dingqi RAO ,Song HUANG ,Yuezhao WANG ,Zhaobin SONG,Xianguang GUO＊ and Jianping JIANG＊

1 Chengdu Institute of Biology,Chinese Academy of Sciences,Chengdu 610041,Sichuan,China

2 Key Laboratory of Bio-resources and Eco-environment of Ministry of Education,College of Life Sciences,Sichuan University,Chengdu 610065,Sichuan,China

3 University of Chinese Academy of Sciences,Beijing 100049,China

4 Jiangsu Key Laboratory for Biodiversity and Biotechnology,College of Life Sciences,Nanjing Normal University,Nanjing 210046,Jiangsu,China

5 State Key Laboratory of Biocontrol/The Museum of Biology,School of Life Sciences,Sun Yat-sen University,Guangzhou 510275,Guangdong,China

6 Kunming Institute of Zoology,Chinese Academy of Sciences,Kunming 650223,Yunnan,China

7 College of Life Sciences,Anhui Normal University,Wuhu 241002,Anhui,China

Abstract From the perspectives of biodiversity conservation and management,there is an urgent need to have at hand current synopses of classification and distributions of species.In this paper,we review and summarize the classifications,Chinese and English names,type specimens,type localities and distributions of China’s lizard fauna to promote scientific exchange and species conservation among relevant people in China and internationally.As of December 31,2020,a total of 230 species of lizard,belonging to 44 genera and 12 families,have been recorded from China,including 4 invasive species:2 in the family Gekkonidae,1 in Iguanidae and one in Dactyloidae.There are 115 endemic species,accounting for 50% of this group.The proportion of endemic species to China was the highest in the family Eublepharidae (84.62%),and the number of endemic species to China was the highest in the family Agamidae (42 species).The species distributions and classification in the“Fauna Sinica (Reptilia 2):Squamata (Lacertilia)”were updated.Among the provincial administrative units,there is a larger number of species in the southern provinces,such as Yunnan,Guangdong,Guangxi and Taiwan.Sichuan,Yunnan,and Taiwan harbour relatively more endemic species than others.A total of 154 species of lizards were first discovered in China,and the type localities of these species are concentrated in Yunnan,Taiwan,Tibet,Xinjiang,and Sichuan.

Keywords classification,type specimen,type locality,distribution,endemic species,invasive species

1.The brief history of lizard bioinventory in China

People’s Republic of China is one of the largest countries in the world (land area about 9.6 million km2),and straddles two zoogeographical realms,the Palaearctic and the Oriental (Zhang,2011).China has a relatively independent geological history and great geological complexity (Hsü and Chen,1999;Wan,2003).For instance,the phased uplift of the Tibetan Plateau caused significant changes in climate,topography and faunal composition beginning in the late Eocene (ca.38 Ma;Favreet al.,2015;Renner,2016).The complex topography in China is generally assigned to four different terraces (Zhang,2011).The terrain is generally high in the west and lower to the east,including the Third Pole,deserts,grassland,mountains,hills,basins,plains,oceans,etc.Tremendous differences in latitude (a span of more than 50° latitude),longitude (a span of more than 60° longitude),and altitude (a span of more than 89 km) create conditions for extremely diverse climates and remarkably heterogeneous landscapes.Thus,a high degree of species richness and endemism in China might have been expected in terms of reptile diversity.

China’s reptiles were recorded sporadically in ancient books.For example,five groups of lizards were recorded in the“Compendium of Materia Medica”,which was completed in 1578 by Shizhen LI (Zhanget al.,1998).These so-called groups are analogous to“families”in the modern classification system.Modern reports on animal surveys in China started as early as the 18th century (Zhanget al.,1998).Prior to 1950,the taxonomic status of orders and suborders was established,and most of the families,genera and species were discovered by overseas scholars (Pope,1935;Zhao and Adler,1993;Zhaoet al.,1999).After 1960,Chinese specialists began to discover new species and establish new higher taxa.With the development of the economy,the increment of government support for field surveys and convenient transportation,the number of new species to science has increased annually,particularly in the 1980s and 2010s (Figure 1).

Figure 1 The trend of new species of lizards discovered by Chinese scientists.

Figure 2 Lizard species diversity in China.

The earliest comprehensive summary of China’s reptile taxonomy is“Retrieval of Chinese Reptile Taxonomy”,which included 117 species,34 genera and 8 families of lizards and their distributions without English names,type specimen and type locality (Herpetological Department,1977).Since then,in 1999,“Fauna Sinica (Reptilia 2):Squamata (Lacertilia)”(Zhaoet al.,1999) was published and included 156 species,39 genera and 9 families of lizards and their distributions without English names and type specimens;this was the most systematic and comprehensive work on China’s lizards to date.In this book,Zhaoet al.(1999) added Eublepharidae and relocated the species of Xenosauridae in China to Shinisauridae.

With the development and progress of molecular systematics,the international classification system of reptiles has also undergone substantial changes,and new taxa have been continuing to be reported.To update recent classification achievements,Caiet al.(2015) catalogued the reptiles in China by reappraising the taxonomic progress made with morphological and molecular data and published "A revised taxonomy for Chinese reptiles",with the addition of the family Sphaerodactylidae.More recently,on the basis of Caiet al.(2015),Wanget al.(2020) summarized amphibians and reptiles (including lizards) in China.These catalogues simply list classification information (Table 1).

Table 1 Comprehensive summary of China’s lizard bioinventory.

Recently,owing to classification changes,cryptic species,and increasing data,the distributions of reptiles in China have changed substantially.Uncertainties in the type specimens and type localities of many species have led to controversies in many classifications.Misidentification also leads to inaccuracies in the distribution data.In the context of increasingly frequent international academic exchanges,the ongoing National Biodiversity Survey and the adjusted List of National Key Protected Wild Animals,the existing literature is scattered and no longer provides the information needed for law enforcement related to species or biodiversity protection to ecological,agricultural,forestry,or other related departments;scientific researchers;or the public.Therefore,from taxonomic and/or resource protection perspectives,there is an urgent need to revise the historical data and compile the classifications and distributions of species.

In this study,the classification,Chinese and English names,type specimen(s),type locality and distribution of China’s lizards were comprehensively summarized for the first time.The distributions in Zhaoet al.(1999) were thoroughly updated,and the changes caused by alterations in species classification were revised.In the end,we completed the latest compilation of information related to 230 known species in China,including 4 invasive species belonging to Gekkonidae,Iguanidae and Dactyloidae.The genus-level and species-level information were updated with discoveries of species,changes in taxonomic status and data from detailed investigations (Table 1 and Section 2).

2.Taxonomic Changes

2.1.PrinciplesThe classification system,Chinese common name,and English common name in the list are based on“(Reptilia 2):Squamata (Lacertilia)”(Zhaoet al.,1999) and“A revised taxonomy for Chinese reptiles”(Caiet al.,2015),with reference to the online checklist of the Reptile Database (www.reptiledatabase.org) and newly published authoritative data as of December 2020.Taken together,a preliminary distributional list of China’s extant Lacertilia reptiles is presented.

In the process of taxonomic reappraisal of published materials,we adhere to the following principles:i) the rules of the International Code on Zoological Nomenclature(“The Code”:ICZN,1999) are correctly applied;ii) monophyly is best,paraphyly is acceptable,and polyphyly is unacceptable;iii) consistency between molecular systematics and morphological classification is ideal;iv) taxa with taxonomic disputes are treated somewhat conservatively to avoid further confusion;and v) taxa should be officially recognized by the scientific community.

The principles of adopting new species and new national records are i) the correct application of“The Code”,ii) the possession of diagnostic characters,iii) adequate documentation of a uniquely designated type species or record of a new specimen,iv) voucher specimens (type specimen or new record specimen) deposited in an internationally accessible biodiversity repository or museum,iv) publication in peer-reviewed scientific journals,v) official recognition by the scientific community,and vi) fieldwork verification.

The principles of adopting Chinese common names are i) to give priority to idiomatic and historical names;ii) consideration of the names commonly used in taxonomy in the past ten years;and iii) scientificity and applicability.

Under these principles,the distributional list was produced by checking numerous historical documents;having many discussions;repeatedly seeking the opinions of domestic and foreign herpetological taxonomists;and removing synonymous,hybrid and misidentified species as well as those not distributed in China.

2.2.GenusWith in the Gekkonidae,the Palaearctic nakedtoed geckos,ranging from North Africa and Central Asia to northern India and western China (Rösler,2017),were reassigned to diverse genera (Baueret al.,2013),such asAlsophylax,Cyrtodactylus,Altiphylax,Cyrtopodion,Tenuidactylus,andMediodactylus.ForCyrtopodionandAltiphylax,according to Bayesian phylogenetic analysis,Cyrtopodion medogenseis considered a member ofAltiphylax(Cheet al.,2020).However,Altiphylaxis not a monophyletic group,and the analysis lacked molecular data forAltiphylax tokobajevi,which is the type species of this genus (Cheet al.,2020).Thus,in support of Zhaoet al.(1999) and Caiet al.(2015),we temporarily keepCyrtopodion sensu latoandretainCy.medogensein this genusto avoid further confusion.In addition,Cyrtodactylus yarkandensiswas considered a synonym ofGymnodactylus stoliczkai(alternative name combination:Altiphylax stoliczkai) (Blanford,1875;Das and Dattagupta,1997;Das,2017) and included inAltiphylax(Baueret al.,2013).However,the only record ofAl.stoliczkaiin China is the type locality ofCy.yarkandensisin Yarkand (=Yarkant County,southern Xinjiang,western China) (Zhao and Alder,1993;Zhaoet al.,1999),which was given by Anderson (1872) and thought to be erroneous (Blanford,1875).The most likely type locality was located at Ladakh (Blanford,1875;Smith,1935;Zhao and Adler,1993;Daset al.,2017).This species is probably not native to China;at least no specimens withbona fidecollection data have been secured at the western Chinese border.For these reasons,Altiphylaxis temporarily not included in our list.Woodet al.(2020) confirmed the placement ofPtychozoontaxa withinGekkowith strong support based on thousands of ultraconserved elements.We adopt their new classification,emphasizing the most inclusive,original generic name (Gekko) for～60 taxa,arranged into seven subgenera.

In the Scincidae,forAsymblepharus,molecular phylogenetics revealed a polyphyletic relationship betweenAsymblepharusandAblepharus,suggesting thatAsymblepharuswas synonymized withAblepharus(Pyronet al.,2013;Cheet al.,2020).Considering the lack of comprehensive studies of these two genera involving molecular and morphological data,we currently recognizeAsymblepharusas a valid genus.The relationships inLygosoma sensu latoare more complex.Freitaset al.(2019) revised the taxonomy by restrictingLygosomato Southeast Asia,resurrecting the genusRiopafor a clade of Indian and Southeast Asian species,expanding the genusMochlusto include all African species ofLepidothyrisand describing a new genus,Subdoluseps,in Southeast Asia.Similar to traditional morphological approaches,multivariate approaches have largely failed to differentiate clades inLygosoma sensu lato(Freitaset al.,2019).Hence,to avoid further confusion,we kept the original generic name (Lygosoma)andtemporarily downgraded these four genera to four subgenera.

In the Lacertidae,forEremias,several 20thcentury herpetologists treatedLacerta veloxPallas,1771 (=Eremias velox),as the type species;this was also adopted by Szczerbak (1974).Subsequently,ICZN used its plenary powers to designateLacerta veloxPallas,1771,as the type species ofEremias(see Melville,1985).

In the Agamidae,forLaudakia,Paralaudakiawas described and separated fromLaudakiaby morphology-based taxonomic revision (Baiget al.,2012) in consideration of mtDNA evidence of a paraphyly ofLaudakia sensu lato.However,owing to a lack of nuclear DNA evidence,the division ofLaudakiainto three genera remains somewhat controversial (e.g.,Pyronet al.,2013),and we temporarily downgraded these two genera to two subgenera and maintained the original generic name (Laudakia).ForCalotes,Oriocalotespaulusis the type species of the genusOriocalotesGünther 1864,which was synonymized withCalotesby Giriet al.(2019).On the basis of combined mtDNA and nuclear loci analyses,Wanget al.(2019) revised the taxonomy ofJapaluraandDiplodermaand splitJapalura sensu latointo four genera,resurrectingDiplodermaand placing it into clade O (with its type species beingDi.polygonatumHallowell,1861),and definedJapalura sensu strictoaccording to clade C (distributed along the southern foothills of the Himalayas).However,by using the mitochondrial ND1-ND2 gene fragments to construct a Bayesian phylogenetic tree,Cheet al.(2020) did not find evidence to support a monophyly of“Japalura sensu stricto”.Considering that the results of Cheet al.(2020) lack nuclear DNA evidence,we temporarily adopt the conclusion of Wanget al.(2019).As suggested by Liuet al.(2020),the Chinese name“ 龙 蜥”refers to the name of the genusJapaluraand the Chinese name“攀蜥”forDiploderma,showing our respect to Professor Chengchao LIU and Ermi ZHAO’s revision.Owing to the morphological similarities,Mahony (2010) synonymized

MictopholistoPseudocalotes.However,theMictopholisspecies was also similar to members of the genusJapalura(Mahony,2010;Wanget al.,2019c).According to phylogenetic studies based on mitochondrial DNA data,Cheet al.(2020) showed thatMictopholiswas more closely related toJapalurathan toPseudocalotes.In support of this perspective,Gowandeet al.(2021),using mtDNA (16S rRNA,ND2 and ND4) and the nuclear RAG1 gene sequences,inferred thatP.austenianawas embedded within the genusJapaluraGray,1853 sensu stricto.

For the gekkonid genusHemidactylus,the authorship is sometimes attributed to G.Cuvier ("1817" [1816]),but he used the form "Hemidactyles",which is a vernacular rather than a Latin form and,thus,unavailable under "The Code".Oken (1817) was the first to use a Latinized form,based on Cuvier's name,by listing the abbreviation "Hemidact." in a list of gecko genera beginning withThecadactylus(fideZhao and Adler,1993).Other systems have credited Grey,1825,without explanation (GBIF Secretariat,2019).The (sub)genusMediodactyluswas first described by Szczerbak and Golubev in 1977,not by Steindachner in 1870.

2.3.SpeciesIn the Eublepharidae,forGoniurosaurus,Go.lichtenfelderiwas recognized as the only species ofGoniurosaurusin China (Zhao and Adler,1993;Zhaoet al.,1999).To date,however,different populations in China have been recognized as distinct species andGo.lichtenfelderiwas found only in Vietnam.Goniurosaurus kadoorieorumis considered a junior synonym ofG.luii,owing to the lack of diagnostic characters separating fromG.luii(Ngoet al.,2016) and a polyphyleticG.luiiwith respect toG.kadoorieorumin both the ML and BI analyses (Grismer,2021).

In the Gekkonidae,forHemiphyllodactylus,the Hunan population ofH.yunnanensis(Denget al.,1998;Shenet al.,2014) was considered a new species(H.dupanglingensis)(Zhanget al.,2020).However,Hemiphyllodactylus dupanglingensisandH.hongkongensismay be synonymous withH.dushanensis(Zhou and Liu,1981) (Yingyong WANG and colleagues,unpublished data).Pending results of analyses with morphological and molecular data published in peer-reviewed scientific journals,we temporarily keep the original distinctions.The Zayü population ofH.yunnanensis(Shiet al.,2011) was also treated as a new species,H.zayuensis(Jianget al.in Cheet al.,2020:439).ForCyrtodactylus,two subspecies ofCy.khasiensiswere found in China,i.e.,Cy.k.khasiensisandCy.k.cayuensis(Liet al.,2010).The Zayü population was the only definite group ofCy.k.cayuensisand was considered a valid species,Cy.cayuensis(Agarwalet al.,2018).The other populations,Mêdog and Longchuan,are known asCy.k.khasiensis(Yang and Rao,2008;Liet al.,2010),which has not been explicitly negated,although Agarwalet al.(2018) stated that“this species is known with certainty only from the vicinity of Sohra,…,Northeast India”.Therefore,we keepCy.khasiensisin our list temporarily.

Inthe Scincidae,forLygosoma sensu lato,onlyLy.bowringii(=Eumeces bowringiiGünther,1864)was recorded in Hong Kong,China.However,the only record in China was questioned:the species has not been recorded subsequently in Hong Kong or other places in China (Smith,1935;Pope,1935;Zhao and Adler,1993;Zhao,1997;Zhaoet al.,1999).Considering that Hong Kong is the type locality ofLy.bowringii,some studies stillincluded this species (Zhao and Adler,1993;Zhao,1997;Zhaoet al.,1999).Until this location is shown to be an error,we keep this species in this fauna of China.

In the Lacertidae,forEremias,it is increasingly common to acceptEr.multiocellatain its entire range,representing a species complex (Eremchenkoet al.,1992;Eremchenko and Panfilov,1999;Sindaco and Eremchenko,2008;Orlovaet al.,2017;alsofideJinglong LIU’s PhD thesis,2019,and Xianguang GUO,as one of his supervisors).Pending the publication of this finding in peer-reviewed scientific journals,we recognize this species asEr.multiocellatatemporarily.Eremias quadrifronsis one of the most enigmatic Central AsianEremias species,known only from a single male specimen (holotype);despite numerous attempts,other specimens of this species were not recorded during subsequent surveys (Szczerbak,1974).Szczerbak (1974) suggested that there is a possibility that the description ofEr.quadrifronswas based on an abnormal specimen ofEr.multiocellata,a thought that was adopted by Zhao (1995) and Zhaoet al.(1999).Xianguang GUO examined the type specimen and considered its morphology normal and not irregular.Further studies,including extensive surveys from the vicinity of this species type locality,are required to clarify its taxonomic status and its phylogenetic position.Eremias roborowskiiwas recommended asEr.velox roborowskiiby Zhaoet al.(1999);recent morphological,ecological and molecular evidence congruently support that it should be raised to the status of full species (Liuet al.,2019;Chirikovaet al.,2019).ForTakydromus,molecular data supportedTa.kuehnei carinatusGressitt,1938 as a full species,despite the lack of a morphological comparison (fideYu’s doctoral thesis,2014,and Xiang Ji as the supervisor).Thus,pending the publication of the results of analyses with morphological and molecular data in peer-reviewed scientific journals,we recognize this taxon as a subspecies.

In the Varanidae,forVaranus,Va.vietnamensisYang and Liu,1994,which is synonymous withVa.nebulosus(Grey,1831) (Böhme and Ziegler,1997),was first described as a species according to a specimen purchased in Hekou County (Rao and Yang,1996).Considering that the specimen was bought from markets,rather than wild-collected in China,this species is not included in this list temporarily.The western Yunnan population ofVa.bengalensiswas treated as a full species (Va.irrawadicus) on the basis of morphological data (Yang and Li,1987;Yang and Rao,2008).Auffenberg (1994) considered that the morphological characteristics were within the range of those ofVa.bengalensisbengalensis,in agreement with Zhaoet al.(1999).DeLisle (1996) and Böhme and Ziegler (1997) considered this taxon a subspecies (Va.bengalensisirrawadicus).Böhme (2003),Pianka and King (2004) and DeLisle (2009) synonymized it withVa.bengalensisand did not recognize any other subspecies.As such,a broad spectrum of studies is needed,especially molecular and morphological studies ofVa.bengalensisfrom the type locality.

In the Agamidae,forLaudakia,La.stoliczkana altaica(Munkhbayar and Shagdarsuren,1970) can probably be treated as a valid species (fideLifang Peng doctoral thesis,2019,and Song Huang as the supervisor).Pending the publication of this study in peer-reviewed scientific journals,we temporarily recognize it as a subspecies.ForPhrynocephalus,different populations ofPh.guttatusin China have been recognized as separate species.Studies of taxonomic status,ecological features and behaviour of these populations by Dunayev (1989) suggested thatPh.guttatusmelanurusdeserves the status of full species,which is supported by molecular evidence (Melvilleet al.,2009;Dunayevet al.,2020;Solovyevaet al.,2018).Phrynocephalus guttatusalpherakiiwas elevated to the status of full species based on morphological and molecular studies (Ananjevaet al.,2011;Solovyevaet al.,2011;Milto and Barabanov,2012;Solovyevaet al.,2018).Phrynocephalus grumgrzimailoiwas considered invalid and synonymous withPh.melanurusby Barabanov and Ananjeva (2007) and Ananjevaet al.(2011),although no evidence was provided.Thus,Ph.guttatusis temporarily not included in this list.Phrynocephalus axillarisandPh.nasatusare not similar enough to be synonymized,because these two species are clearly distinguished from each other by morphological traits (colouration and scalation) and represent different phylogenetic lineages (Dunayev,2020).Phrynocephalus alpherakiiwas first described in 1907 (Bedriaga,1907a).Phrynocephalus grumgrzimailoiandPh.putjataiwere first described in 1909 (Bedriaga,1909).In the viviparous group ofPhrynocephalus,i.e.,the subgenusOreosaura,Ph.erythrurusis sister toPh.vlangalii,not the synonym ofPh.theobaldi(Jin and Brown,2013;Solovyevaet al.,2018).Bayesian species delimitation analysis of mtDNA did not support the division ofPh.putjataiandPh.guinanensisinto separate species (Jinet al.,2014).Genotyping by sequencing further suggested synonymisingPh.guinanensiswithPh.putjatia(Jin and Brown,2019).However,according to Xiang Ji’s unpublished data,there were significant differences in morphology,genetics of topotypes and life history adaptation to the pure desert environment betweenPh.putjataiandPh.guinanensis,and it is believed thatPh.guinanensisis in the process of ecological species formation.ForDiploderma,Ota (2000) synonymizedJapalura szechwanensisHu and Djao,1966 (=Diploderma szechwanensis),withJ.fasciataMertens,1926 (=Diploderma fasciata).Nevertheless,Ota compared one type specimen ofJ.fasciataandthree type specimens ofJ.szechwanensisby using only colour patterns and nine common quantitative characters.This evidence is not adequate to synonymize the two species.In addition,the biogeographic characteristics of western Sichuan (the type locality ofJ.szechwanensis) and northern Indochina are quite different.Thus,we temporarily recognizeJ.szechwanensisas a valid species.ForPseudocalotes,Yang and Rao (2008) treatedJapalura bapoensis(=Ps.bapoensis)as a full species on the basis of morphological data.Based on molecular data,Wanget al.(2019) considered it a member ofPseudocalotesand conservatively treated it as a subspecies ofPs.kingdonwardi,withoutmolecular data ofPs.kingdonwardifrom the type locality.As such,we temporarily keepPs.bapoensisas a full species.A broad spectrum of studies is needed,especially molecular and morphological studies ofPs.kingdonwardifrom the type locality.

In terms of the authorship and date of species descriptions,forAlsophylax,in accordance with Art.23.9 of "The Code",Al.przewalskiiStrauch,1887,was declared anomen protetcum,with priority over thenomen oblitum Gymnodactylus microtisBlanford,1875.InHemidactylus,He.platyuruswas first described in 1792,not in a reprint in 1797 (Schneider,1792,1797).ForSphenomorphus,thefirst paper with a description ofSp.incognitusisThompson’s,which was apparently published a month earlier than van Denburgh’s (fideZhao and Adler,1993).ForEremias,Eremias buechneriandEr.roborowskiiwere first printed in 1907;this is similar toTeratoscincus roborowskii(Bedriaga,1907b).ForZootoca,Lacerta vivipara(=Z.vivipara) was first used as a“real”species name in Lichtenstein’s (1823) catalogue,but“Lacerta viviparaJacquin,1787”did not become a viable name (Schmidtler and Böhme,2011).

2.4.Type specimen and Type localityWe added some holotypes and field numbers for type specimens (Table 2).The holotype ofGekko taibaiensisis IZSX 840108,not IZSX 840104 (Song,1985).A neotype ofCalotes versicolor(NCBS AT102) was designated by Gowandeet al.(2016) but the action was invalidated by Chaitanyaet al.(2017).

Table 2 Some holotypes and field numbers were added.

We summarized and clarified some type localities (Table 3).The type locality ofSc.przewalskiiis Mt.Dschachar,on the upper Chuanche (probably Bailong Jiang),Gansu Province,China (fideZhao and Adler,1993).The Dschachar Mountains probably refer to Die Mountain (迭山),which are located on the upper end of the Bai-Long River (白龙江).“Dschachar”and“Chuanche”may be pronounced in the Khams Tibetan Zhouqu dialects.

Table 3 Clarified type localities.

The type locality ofSc.schmidticonfused Dr.Ermi ZHAO,who was“not sure if it referred to Dawa Shan (alternative name:Wa Shan) in Jinkouhe District or Wa Shan (alternative name:Wawu Shan) in Hongya County and Yingjing County”(Zhao,2003:117).In historical references,the locations and landforms of Wawu Shan and Wa Shan were described in detail by Ernest H.Wilson (1913),who is the director of the expedition sent out by the Arnold Arboretum and the associate of Walter R.Zappey,who collected the holotype of this species.From Ernest H.Wilson’s book,we can verify that the type locality ofSc.schmidtiis“Dawa Shan (Wa Shan) Mountain”in Jinkouhe District,not“Wawu Shan (Wa-wu Shan)”.

The type locality ofDr.maculatusis Penang,Malaysia (restricted by Smith,1935:138).This type locality may be an error,because Smith (1935:140) claimed that the species is not known south of 8° north latitude (Zhao and Adler,1993).

3.Distributional list and its analysis

3.1.The distributional list of China’s lizardsThe distributional data in Zhaoet al.(1999) were thoroughly updated in this paper,as new records were added and the changes caused by the alteration of species classification were corrected.The detailed changes are listed in Appendix 1 (the distributional list of native and extant lizards in China).

The distributions of invasive lizards are enumerated in Appendix 2 (the distributional list of invasive lizards in China).

3.2.Lizard species diversityAs of December 31,2020,12 families,44 genera and 230 species of lizards (including invasive species) have been found in China.There were 115 endemic species,accounting for 50% of this group.The proportion of endemic species was the highest in Eublepharidae (84.62%),and the number of endemic species was the highest in Agamidae (42 species).There were 4 invasive species:2 in Gekkonidae,1 in Iguanidae and 1 in Dactyloidae.

3.3.Distribution patternFrom the tropics to the poles,as the latitude increases,the species diversity decreases with the increasing latitude,the most typical latitude gradient pattern in the distribution pattern of species diversity (Kimmins,1987;Rosenzweig,1995).This pattern embodies in terrestrial plants,vertebrates,invertebrates,marine life and even ancient creatures (Hawkinset al.,2003;Coxet al.,2019),also in lizards in China (Huanget al.,2011).

In terms of species richness,the Nanling Mountains and tropical rainforests in South China harbour the largest number of species,followed by the Wuyi Mountains in East China and the Hengduan Mountains,Qinling Mountains and Wushan Mountains in Southwest China.The number of species is the lowest in Northeast and North China.Among the provincial administrative units,the number of species in the southern provinces,such as Yunnan,Guangdong,Guangxi and Taiwan,is larger (Figure 3).These provinces are in the tropical and subtropical zones,have more complex terrain,and can support more species.In the Tibetan Plateau and northern or small provincial administrative units,the number of species is smaller.The environment in these provinces cannot support the survival of a large number of species.

Figure 3 The proportion of nonendemic native,endemic and invasive lizard species on an administrative division map of China.Map approval number:GS(2020)3888

The provinces with more complex terrain,which contain many mid-elevation mountains and low-elevation hills,such as Sichuan,Yunnan,and Taiwan,especially in the broader Hengduan Mountains region,harbour more endemic species (Figure 4).

A total of 154 species of lizards were first discovered in China,and the type localities of these species are concentrated in Yunnan,Taiwan,Tibet,Xinjiang,and Sichuan (Figure 4).The type localities of 7 species are too wide to mark to the map.And 7-10 type localities need further restriction.

AcknowlegementsWe thank Natalia B.Ananjeva and Herbert Rösler for providing literatures,and Pipeng LI,Wenge ZHAO and Dajie GONG for confirming species distribution information.We also thank Xuemei DU for help in the production of Figure 4.This research was supported by the Strategic Priority Research Program of the Chinese Academy of Sciences,Grant No.XDA19050201,the National Natural Science Foundation of China (32070433),the Special Fund for Youth Scholars on Taxonomy,the Chinese Academy of Sciences (ZSBR-014)，China Biodiversity Observation Networks (Sino BON -Amphibian &Reptile).

Figure 4 The type localities of nonendemic native and endemic species on a topographic map of China (1.round:type localities of endemic species;triangle:type localities of non-endemic species;2.brown:Dibamidae;red:Eublepharidae;yellow:Sphaerodactylidae;green:Gekkonidae;purple:Scincidae;blue:Lacertidae;pink:Anguidae;white:Shinisauridae;cyan:Agamidae).Map approval number:GS(2020)3888.